G Model
ARTICLE IN PRESS
NBR 1952 1–7
Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Neuroscience and Biobehavioral Reviews journal homepage: www.elsevier.com/locate/neubiorev
Review
1
Lateralized perception: The role of attention in spatial relation processing
2
3
4 5 6 7
Q1
Ineke J.M. van der Ham a,∗ , Albert Postma a,b , Bruno Laeng c a b c
Department of Experimental Psychology, Utrecht University, The Netherlands Department of Neurology, Utrecht University Medical Center, The Netherlands Department of Psychology, University of Oslo, Norway
8
9 20
a r t i c l e
i n f o
a b s t r a c t
10 11 12 13 14
Article history: Received 16 December 2013 Received in revised form 8 May 2014 Accepted 13 May 2014
15
19
Keywords: Spatial relation processing Hemispheric lateralization Spatial attention
21
Contents
16 17 18
22 23 24 25 26 27 28 29 30
1. 2. 3. 4. 5. 6. 7.
Any spatial situation can be approached either categorically – the window is to my left – or coordinately – the glass is 20 cm away from the bottle. Since the first description of the distinction between categorical and coordinate spatial relation processing, it has often been shown that they are processed by at least partially different underlying mechanisms, mainly located in the left and right hemisphere, respectively. A number of recent studies have suggested that spatial attention plays a particularly important part in the perception of space: categorical processing benefits from a local focus of attention, and coordinate processing profits from a global focus of attention. This suggests that the lateralization pattern is modified by the concurrent size of the attentional focus, and is consequently more dynamic than previously thought. Therefore, a thorough revision of earlier theories on spatial relation processing is in order. In this review, we present a new model on lateralization of spatial relation processing that explicitly describes the role of spatial attention. © 2014 Published by Elsevier Ltd.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Empirical evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Critical evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Attentional scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bringing together dichotomies in low-level perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Generalizing the dichotomy to higher level perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
00 00 00 00 00 00 00 00 00
31 39 32
33 34 35 36 37 38
1. Introduction When we interact with our environment, its spatial features are crucial to us. There are different ways to deal with these spatial features. If someone asks us where we left the car keys, we can respond with something like: ‘I left them on the coffee table in the living room’. On the other hand, when we want to pick up those keys, it does not help much to know they are on the table; we need
∗ Corresponding author at: Heidelberglaan 1, Room H0.13, 3584 CS Utrecht, The Netherlands. Tel.: +31 030 2534023; fax: +31 030 2534511. E-mail address: [email protected] (I.J.M. van der Ham).
to know their exact location with respect to our hand in order to plan and execute our movement. These two examples illustrate the two main types of spatial relation processing. The first example shows that we can process spatial relations between objects in an abstract, propositional way, which is termed categorical. We can use this type of spatial relation to give directions or memorize where we left things. In contrast, in the second example we use the precise, metric properties of spatial relations, or coordinate spatial relations, when we navigate or grasp objects. The characteristics of these spatial relations are in part dictated by the type of ‘coordinate system’ that is used to represent spatial characteristics. Such coordinate systems can be either egocentric (observer based) or allocentric (environment based) (see
http://dx.doi.org/10.1016/j.neubiorev.2014.05.006 0149-7634/© 2014 Published by Elsevier Ltd.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
40 41 42 43 44 45 46 47 48 49 50 51 52
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
2
e.g. Klatzky, 1998; Majid et al., 2004). In an egocentric coordinate system, these spatial relations by definition concern the relation between an external object and the observer, such as ‘the car is to 55 my left’ or ‘I am closer to the library than you are’. Neurophysiolog56 ical studies on primates indicate that visual space is constructed 57 many times over, according to a variety of frames of reference, 58 each attached to different parts of the body and possibly subserv59 60Q2 ing different functions (Graziano and Gross, 1995). Neuroimaging in humans confirms that one of the topographic maps of the pari61 etal lobes is based on a head-centered coordinate frame (Sereno 62 and Huang, 2006). A plurality of sensorimotor action spaces can be 63 related to specific effectors that have the ability to move indepen64 dently from the rest of the body (e.g. hand, head, and eye). In such 65 motor-oriented frames of reference, spatial relationships between 66 two locations can be coded in terms of the movements required to 67 act or move from one to the other (Paillard, 1991). 68 In an allocentric coordinate system the relation concerns two 69 objects or parts of objects that are defined according to a refer70 ence frame whose fulcrum is located outside of the observer‘s body. 71 Examples of spatial relations from an allocentric perspective are: 72 ‘the church is to the North of the square’ and ‘you are 200 meters 73 West from the library’. Crucially, both categorical and coordinate 74 spatial relations can be applied within these different frame of ref75 erence and coordinate systems (see also Ruotolo et al., 2011a,b; 76 Ruggiero et al., 2012). In this review we focus on the direct com77 parison of categorical and coordinate spatial relations, regardless 78 of the coordinate system used. All that is discussed here can equally 79 apply to spatial relation processing in an egocentric as well as an 80 allocentric or object-based perspective. 81 The distinction between categorical and coordinate spatial rela82 tions is not only reflected by clear differences in functionality, 83 but also refers to different underlying processing mechanisms. 84 Although described earlier by McNamara (1986), Kosslyn (1987) 85 provides the first elaborate theoretical framework to understand 86 the differences between categorical and coordinate processing. He 87 introduced the idea that the two types of spatial relations are pro88 cessed by two different subsystems in the human brain, which have 89 lateralized due to evolutionary processes. Categorical spatial rela90 tion processing is thought to be lateralized to the left, whereas 91 coordinate spatial relation processing is argued to be right later92 alized. 93 Over the years, many experimental studies have been performed 94 concerning this distinction, which have enabled a fine-tuning of the 95 original theory (see for review, e.g. Jager and Postma, 2003; Laeng 96 et al., 2003; Laeng, 2014). Recent empirical studies have brought 97 new evidence to the table and point toward a substantial role of 98 spatial attention in explaining differences between categorical and 99 coordinate processing. This offers a novel theoretical perspective. 100 Therefore, an up-to-date review of current evidence and the impact 101 of these findings on the theoretical framework underlying cate102 gorical and coordinate spatial relation processing is called for. In 103 this review, we first provide an overview of the main experimental 104 work on spatial relation processing, particularly recent work. Then 105 we discuss what this tells us about the nature of spatial relation 106 processing, and focus on the role of spatial attention in particu107 lar. We present a new framework to fit the experimental findings 108 on how spatial attention affects spatial relation processing and its 109 lateralization. Lastly, we will explore how spatial relation theory 110 translates to human visual perception in general. 111 53 54
112
113 114 115
2. Empirical evidence Soon after Kosslyn’s first publication on the topic of spatial relation processing (Kosslyn, 1987) numerous behavioral studies attempted to empirically prove the distinction (e.g. Hellige and
Michimata, 1989; Kosslyn et al., 1989; Koenig et al., 1990; Bruyer et al., 1997; Wilkinson and Donnelly, 1999; Banich and Federmeier, 1999). These studies were mainly aimed at verifying the double dissociation of type of spatial relation and hemisphere. Typically, very simple stimuli were used in these studies. The dot-bar stimuli have been used in most of these early studies. These stimuli consist of a horizontal bar with a dot presented either above or below the bar, at varying distances, as illustrated in Fig. 1. The categorical instruction in this case, was to indicate whether the dot is ‘above’ or ‘below’ the bar, regardless of distance. The coordinate instruction in turn, was to focus on the distance and indicate whether the dot is ‘within’ or ‘not within’ a predetermined distance, regardless of side. An important feature of this design was that the stimulus layout was identical and only the instruction varied between the two versions of the task. In these behavioral studies, a visual half field design was used to assess lateralization patterns by means of response times and accuracy for stimuli presented to the left or right visual field. This approach was soon followed by studies in which lateralization was determined with more direct and precise measures of brain activity: positron emission tomography (PET), functional magnetic resonance, imaging (fMRI and MEG), electroencephalography (EEG), and transcranial magnetic stimulation (TMS). In the vast majority of these studies, the lateralization pattern was further substantiated (e.g. Baciu et al., 1999; Kosslyn et al., 1998; Trojano et al., 2002, 2006; van der Ham et al., 2009; Franciotti et al., Q3 2013). Several studies on patients with unilateral brain damage also supported these findings (e.g. Laeng, 1994; Palermo et al., 2008; van der Ham et al., 2011, 2012a,b,c). Overall, the parietal cortex Q4 seemed of particular importance; left parietal cortex was mainly linked to categorical processing, whereas the right parietal cortex was involved mostly with coordinate processing. Yet, the original theory as proposed by Kosslyn needed some attenuation. Most of the foregoing findings indicate that the lateralization pattern is not mutually exclusive: both hemispheres are involved in processing each type of spatial relations, but they show a clear bias toward one of the two types. Aside from the many techniques that have been used to study this dichotomy, various task designs have been applied as well, facilitating different research questions. The first number of studies was designed to study spatial relation processing in perception: a response was required to a briefly presented, simple stimulus. Results from these studies illustrate how we process the spatial features of visual stimuli. Later experiments also focused on spatial relation processing in working memory task designs, in which two sequentially presented stimuli were compared (e.g. Laeng, 1994; van der Ham et al., 2007), and in a mental imagery paradigm (Michimata, 1997; Palermo et al., 2008) where clock faces had to be imagined. These studies have shown that the typical lateralization pattern reflecting separate processing mechanisms can also be generalized to visual working memory and mental imagery. Although the double dissociation of relation type and hemisphere is found in the convincing majority of these studies, the effects have shown to be sensitive to the methodology used. For instance, Baciu et al. (1999) showed that over time, participants started to categorize the coordinate task, as reflected by a shift from a right hemisphere to a left hemisphere dominance during the experiment. This has been explained with a critical and disadvantageous feature of the dot-bar stimulus layout. The dot could only appear at a fixed number of positions and the instruction entailed a categorization of distances into ‘within’ a particular distance and ‘not within’ a particular distance. Therefore, participants could well start to realize this limitation and categorize the distances instead of encoding them as exact distances. The cross-dot working memory paradigm (see Fig. 2) in which two distances were compared later solved this problem as participants continued to encode the
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
3
Fig. 1. The often used dot-bar stimulus, (A) with all possible dot positions, (B) dots above the bar belonged to one spatial category, the dots below the bar to the other spatial category, and dots within a particular distance of the bar belonged to one type of coordinate relation, dots not within this distance belonged to the other type of coordinate relation. Note the during the task, only one dot position would be visible.
182 183 184 185 186 187 188 189 190 191 192
exact distance, not a distance category in this type of task. Also the higher amount of possible dot positions avoided categorization. Another study (van der Ham et al., 2007) showed that categorical information decays slower than coordinate information. In a working memory task design, the interval between two stimuli therefore determines whether a double dissociation is found or not, typically it is found with an interval of 500–2000 ms. In short, a large number of studies making use of various techniques and task designs demonstrate that there is a relative advantage of the left hemisphere in processing categorical information, and of the right hemisphere in processing coordinate
information. Yet, specific task demands can affect this lateralization pattern. 3. Critical evaluation A main point of criticism that can be made is whether it is really meaningful to study this distinction. What could we learn from such a distinction? Newell (1973) criticizes the tendency of psychologists to test binary distinctions, as supposedly these are typically not present in nature. Kosslyn (2006) argues against this statement, as for spatial relation processing such a clear distinction
Fig. 2. Version of the cross dot stimulus, used in working memory designs, (A) with all possible dot positions, (B) with a given dot position for the first stimulus (S1), the dots shown as S2 that would match categorically are shown on the left, and the dots that would match coordinately are shown to the right. Note that during the task, only one dot position would be visible for both S1 and S2.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
193 194
195
196 197 198 199 200 201
G Model NBR 1952 1–7 4
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
is present. In fact, the general idea behind lateralization of spatial relation processing should be viewed in a more general way. The brain is thought to ‘divide-and-conquer’ when it comes to processing information. If incoming information needs to be processed, it would be inefficient if both hemispheres would do this in the same way and then needed to communicate about this. Therefore, from an evolutionary point of view, lateralization, or the division of labor between the two hemispheres, will lead to more efficient and faster processing (see e.g. Hugdahl, 2000; Cook, 1986). Although categorical and coordinate spatial relations are generally thought to be based on partially separate processing mechanisms, sufficient evidence exists for it to be considered a clear binary, and lateralized distinction. Furthermore, we would like to argue that studying how we process spatial relations is meaningful, as it can be generalized to many areas of perception, as discussed later. Despite the large amount of supporting evidence for the dichotomy, alternative explanations have been proposed as well. One prevalent alternative is that of space versus language. The traditional line of reasoning is that the right hemisphere typically processes spatial information, whereas verbal information processing is located in the left hemisphere. Therefore, it has been argued that in particular the left hemisphere advantage of categorical processing is due to the linguistic properties of this type of processing. In the dot bar task for instance, the use of propositions ‘above’ and ‘below’ suffices to solve the task. Kemmerer and Tranel (2000) argued there might be a triad instead of a dichotomy, with categorical relations split up into strictly spatial categories and verbal categories. They proposed that only the processing of verbal categories shows a left hemisphere bias. Yet, a number of studies argue against the fact that left lateralization is based primarily on the verbal content of processing. For categorical perception in general, it has been shown that preverbal infants show mainly left hemisphere activity when they categorize spatial orientations (Franklin et al., 2010). Also, no difference in lateralization pattern is found when people use labeled or unlabeled categories, indicating that language is not the main factor in the direction of lateralization (Holmes and Wolff, 2012). Within spatial relation processing in particular, the verbal nature of categorical stimuli could only explain for a minor part of the left lateralization (van der Ham and Postma, 2010). This is further substantiated by Suegami and Laeng (2013), who found that although categorical relations can be coded semantically as well as visuospatially, a left lateralization is found for both. As language processing is convincingly based in the left hemisphere, it is likely that it has some effect on the extent of lateralization when participants use verbal strategies to solve categorical tasks. Yet, given the current evidence it seems highly unlikely that language by itself is the determining factor in the direction of lateralization. A second alternative explanation that has come up is that of difficulty. Typically, processing categorical relations is easier to do than processing coordinate relations. Sergent (1991a,b) was the first to therefore suggest that instead of a dichotomy of categorical and coordinate, it might be a continuum between easy and difficult. The more difficult a task would get, the more the right hemisphere would get involved. Although Sergent’s point of view was experimentally disproven by Kosslyn et al. (1992) and Slotnick et al. (2001), this point has also been used to explain experimental results by Van der Lubbe et al. (2006) and Martin et al. (2008) in later studies. Yet, a more recent fMRI study van der Ham et al. (2012a,b,c) again demonstrated that task difficulty itself cannot explain for the lateralization patterns found (see also Franciotti et al., 2013). It cannot be denied that there are adherent difficulty differences between categorical and coordinate processing, but when reviewing the evidence, difficulty does not seem to determine the lateralization pattern.
4. Attentional scope Given that categorical and coordinate spatial relation processing are markedly distinct at a neurocognitive level and cannot be merely explained by characteristics like involvement of language or task difficulty, the question remains what exactly is the nature of spatial relation processing? A number of recent studies suggest a novel, appealing answer to this question, by looking into spatial attention and more specifically the size of attentional scope. Spatial attention is not an entirely new element in the ideas on spatial relation processing. Actually, in his first report Kosslyn (1987) already commented on the issue. He described the purpose of spatial relation interpretation as looking up the location of the relevant part and directing attention to this location, leading to the output of instructing the attention-shift subsystem to shift attention accordingly. He proposed that for both categorical and coordinate interpretation this process is the same, except for the nature of the representation used to do this, which is either categorical or coordinate. Several behavioral studies indicate that the difference in representation – categorical or coordinate – is closely related to the size of the attended location. Categorical processing is found to benefit most from a local, small focus, whereas coordinate processing seems more efficient with a global, large focus of attention. In several experiments this same pattern was found when manipulating the area participants attended to (Okubo et al., 2010; Michimata et al., 2011; Borst and Kosslyn, 2010; Laeng et al., 2011; Franconeri et al., 2012; Franciotti et al., 2013). Moreover, the same attentional pattern was also found in the absence of experimental manipulations. In a case study, patient NC showed great difficulty in processing coordinate information in particular, which coincided with a clear local bias, as opposed to the global bias generally found in human visual perception (van der Ham et al., 2012a,b,c). In line with this, it has also been shown that children with right hemisphere damage show a combination of impairment in global focus and coordinate judgments (Schatz et al., 2004). Furthermore, with retinotopic mapping of fMRI data, it has been shown that participants directed their attention more locally when they were asked to give categorical responses, as compared to coordinate responses (van der Ham et al., 2012a,b,c). The interpretation of this dissociation between categorical and coordinate processing in terms of size of attentional focus revives an older idea about spatial relation processing: the link to receptive field sizes. In a computational model (Kosslyn et al., 1992) the flow of information in processing categorical and coordinate information was simulated. This simulation showed that the categorical decisions benefited from the use of small receptive fields, whereas the opposite was found for coordinate information. This difference makes sense when considering the functionality of the two types of relations. The smaller, non-overlapping fields facilitate splitting up space into distinct areas, or categories, and large, overlapping fields facilitate ‘coarse coding’ (Hinton, 1981), which helps to derive precise metric information. This is substantiated by the studies showing that attentional focus is different between categorical and coordinate processing. An important addition provided by the spatial attention studies is that they show that it concerns a dynamic process: the size of the attentional focus seems to be attenuated according to the task at hand, rather than to be fixed (Fig. 3). Based on the recent empirical evidence concerning the role of attentional focus, we present a new model to explain for the lateralization effects found for spatial relation processing. This model highlights that both task instruction and stimulus size have a pivotal role in predicting the lateralization effect. Task instruction reflects the original theory: a categorical task instruction will lead to a left hemisphere advantage, and a coordinate instruction to a right hemisphere advantage. Stimulus size has a major role in this
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
268
269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
Fig. 3. The model of information processing resulting in a spatial relation decision. Both task instruction and stimulus size contribute to the lateralization pattern.
359
model. It refers to the size of the crucial elements of the visual stimulus and directly determines the attentional focus that is required for a specific spatial relations decision. When stimulus size is relatively small, the attentional focus will also be relatively small and lead to a left hemisphere advantage in processing. Conversely, a relatively large stimulus size will create a large area of attentional focus and consequently result in a right hemisphere advantage. This new model highlights that lateralization in visual perception is not solely determined by task instruction. Attentional focus has been demonstrated to yield a considerable impact as well. The addition of stimulus size not only explains for the recent findings concerning small and large attentional scope and categorical and coordinate spatial processing, respectively, it could also account for variations in extent of lateralization found in different studies over the years. Precise stimulus features like size have been largely ignored over the years. The main criterion has been that the stimulus layout would be identical for categorical and coordinate decisions, in the sense that only the instruction would determine whether it was a categorical or coordinate task. Yet, the elements of those stimuli that are crucial for making a correct decision differ nonetheless. For instance, with the cross dot stimuli in Fig. 2, the categorical decision only requires to check one fourth of the total stimulus area (the second dot is either in that area or not), whereas for the coordinate decision, the entire stimulus area should be attended (the second dot can appear in any direction relative to the center). Therefore, it is important to realize there are two forces at work here.
360
5. Bringing together dichotomies in low-level perception
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358
5
spatial frequency hypothesis; the left hemisphere is more proficient in processing high frequency and the right hemisphere in processing low frequency information (Sergent, 1983). This has been supported by a number of studies (e.g. Mecacci, 1993; Proverbio et al., 1997). Surprisingly little research has been done in which these different dichotomies have been directly integrated. Frequency manipulations have been applied in categorical-coordinate tasks to investigate a possible interaction. Okubo and Michimata illustrated that the coordinate right hemisphere advantage disappeared when low spatial frequencies were eliminated from the stimuli (Okubo and Michimata, 2002, 2004). The categorical left hemisphere advantage disappeared when high spatial frequencies were filtered out in the stimuli. Yet, such a relation between spatial frequency and spatial relation processing is contradicted by Niebauer and Christman (1999). The ‘double filtering by frequency model’ created by Ivry and Q5 Robertson (1998) provides a theoretical framework for this particular topic. In their model they provide a unitary explanation of lateralization effects in visual perception. They base their claim on the premise that some of our perceptual systems make use of a Fourier analysis to describe incoming information, and that the left and right hemisphere have a bias toward processing high and low spatial frequencies, respectively. Their model is therefore based on the frequency characteristics of incoming stimuli. It proposes three separate and sequential stages of stimulus processing: sensory representation, selective filtering of task-relevant information, and asymmetric filtering by cerebral hemisphere of selected information. In the first filtering stage an attentional mechanism serves as a filtering system with regard to the frequency spectrum; the region selected in the spectrum can be as narrow or broad as required by the task performed. The actual lateralization does not occur until the second filtering stage, during which higher order analyses are taking place, for which each hemisphere has its preference. The processing in the left and right hemisphere is considered high-pass and low-pass filtering operations, respectively. As Ivry and Robertson (1998) state: “The central hypothesis of the DFF theory is that the output of those filtering operations underlies many of the laterality effects reported in visual and auditory perception”. These laterality effects are very likely to include local-global and categorical-coordinate processing. The authors indicate the model is a post hoc interpretation of experimental studies and new empirical input is necessary to examine its validity. The recent findings concerning attentional focus, described above appear to fulfill this need. As the local and global focus of spatial attention is directly related to categorical and coordinate spatial relation processing, respectively, these findings underline this model. Importantly, ‘task instruction’ as defined in the model does not seem to be restricted to spatial relation task instruction, but can easily be replaced by other types of spatial instructions, like local versus global.
6. Generalizing the dichotomy to higher level perception 361 362 363 364 365 366 367 368 369 370 371 372 373 374
By discussing size of attentional focus, the local versus global dichotomy naturally comes to mind. This is another classical dichotomy proposed to describe human perception (e.g. Navon, 1977) that has been shown to be lateralized (e.g. Martin, 1979). Although it is an inherent continuous property, it is typically regarded as a dichotomy of ‘relatively small’ versus ‘relatively large’. A meta-analysis confirmed the general finding that there is left hemispheric specialization for local information and the right hemisphere shows an advantage in processing global information (Van Kleeck, 1989). Furthermore, the notion of receptive field sizes is directly related to the processing of high and low spatial frequencies, also a dichotomy with a distinct lateralization pattern. The commonly found direction of lateralization in spatial frequency processing tasks has been framed in the
With a new model reflecting the main findings concerning spatial relation processing, it is imperative to attempt to explore generalization of this theory. As mentioned earlier, the vast majority of fundamental studies providing information about how we process categorical and coordinate information use very simple stimulus layouts, consisting of lines, crosses, and dots. How do these findings relate to how we process visual information in real life? Several researchers have asked this question and looked into spatial relation processing within objects (e.g. Laeng et al., 1999; Saneyoshi et al., 2006; Saneyoshi and Michimata, 2009), faces (Cooper and Wojan, 2000), natural scenes (van der Ham et al., 2012a,b,c), and even in the dynamic process of navigation (Baumann et al., 2012). These studies have largely underlined the evidence from the
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423
424
425 426 427 428 429 430 431 432 433 434 435 436 437
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
6
studies with simple stimuli: a clear dissociation between the two types in terms of underlying processes. So, the distinction appears to hold regardless of visual complexity, meaning, and context. Yet, 440 Q6 in a study on spatial relations in objects Amorapanth et al. (2010) 441 apply some nuance to this: only part of the brain areas involved 442 in spatial relation processing were separate, other areas appeared 443 to be shared for both types. Taken together, it seems that the way 444 in which we perceive the spatial relation information in the world 445 finds its basis in the theories discussed here. 446 438 439
447
7. Conclusion
476
Over 25 years of research supports the view that categorical and coordinate spatial relations are processed by at least partially different underlying mechanisms, mainly located in the left and right hemisphere, respectively. Moreover, this distinction is found within perception (of either space or objects), working memory, and mental imagery. Several criticisms have been uttered in response to these findings. Yet, the alternative explanations for the findings recapitulated here do not match up to the existing theory: language only seems to play a minor part and a difficulty continuum seems highly unlikely. Spatial attention plays a particularly important part in recent empirical studies on the perception of space. Categorical processing benefits from a local focus of attention, whereas coordinate processing is more efficient when a global focus of attention is applied. Therefore, the original theory on the role of receptive field sizes has been revived, and seems to be more flexible than previously thought. Furthermore, it also highlights the interplay between various dichotomies widely described within perception and in part supports Ivry and Robertson’s Double Filtering by Frequency Model. We propose that the clear distinction between categorical and coordinate is closely linked to the continuous measure of attentional scope. The rigid distinction between the two can therefore be attenuated by this factor, as a result of experimental characteristics and innate preferences. Finally, we see that the findings discussed here do not only reflect how we process simple laboratory stimuli, but also how we see the world. The dichotomy has shown to be a major factor as well in how we process objects, faces, scenes, and interact with our environment.
477 Q7
Uncited references
448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475
478 479 480
481
482 Q8 483
484
485 486 487 488 489 490 491 492 493 494 495 496 497
Dent (2009), Kitterle et al. (1990), Kosslyn and Jacobs (1994), Marr (1982), Robertson and Lamb (1991), van der Ham and Borst (2011a,b) and van der Ham et al. (2010). Acknowledgment IH was supported by a Veni grant (451-12-004) by NWO (the Netherlands Organisation of Scientific Research). References Amorapanth, P., Widick, P., Chatterjee, A., 2010. The neural basis for spatial relations. J. Cogn. Neurosci. 22 (8), 1739–1753. Baciu, M., Koenig, O., Vernier, M.-P., Bedoin, N., Rubin, C., Segebarth, C., 1999. Categorical and coordinate spatial relations: fMRI evidence for hemispheric specialization. Neuroreport 10, 1373–1378. Banich, M.T., Federmeier, K.D., 1999. Categorical and metric spatial processes distinguished by task demands and practice. J. Cogn. Neurosci. 11, 153–166. Baumann, O., Chan, E., Mattingley, J.B., 2012. Distinct neural networks underlie encoding of categorical and coordinate spatial relations during active navigation. Neuroimage 60 (3), 1630–1637. Borst, G., Kosslyn, S.M., 2010. Varying the scope of attention alters the encoding of categorical and coordinate spatial relations. Neuropsychologia 48 (9), 2769–2772.
Bruyer, R., Scailquin, J.-C., Coibion, P., 1997. Dissociation between categorical and coordinate spatial computations: modulation by the cerebral hemispheres, task properties, mode of response, and age. Brain Cogn. 33, 245–277. Cook, N.D., 1986. The Brain Code: Mechanisms of Information Transfer and the Role of the Corpus Callosum. Methuen, London. Cooper, E.E., Wojan, T.J., 2000. Differences in the coding of spatial relations in face identification and basic-level object recognition. J. Exp. Psychol. Learn. 26, 470–488. Dent, K., 2009. Coding categorical and coordinate spatial relations in visual–spatial short-term memory. Q. J. Exp. Psychol. 62 (12), 2372–2387. Franciotti, R., D’Ascenzo, S., Di Domenico, A., Tommasi, L., Laeng, B., 2013. Focusing narrowly or broadly attention when judging categorical and coordinate spatial relations: a MEG study. PLoS ONE 8 (12), e83434. Franconeri, S.L., Scimeca, J.M., Roth, J.C., Helseth, S.A., Kahn, L.E., 2012. Flexible visual processing of spatial relationships. Cognition 122, 210–227. Franklin, A., Catherwood, D., Alvarez, J., Axelsson, E., 2010. Hemispheric asymmetries in categorical perception of orientation in infants and adults. Neuropsychologia 48, 2648–2657. Graziano, M.S.A., Gross, C.G., 1995. Multiple representations of space in the brain. Neuroscientist 1, 43–50. Hellige, J.B., Michimata, C., 1989. Categorization versus distance: hemispheric differences for processing spatial information. Mem. Cognit. 17, 770–776. Hinton, G.E., 1981. Shape representation in parallel systems. In: Drina, A. (Ed.), Proceedings of the Seventh International Joint Conference on Artificial Intelligence. MIT Press, Cambridge, MA. Holmes, K.J., Wolff, P., 2012. Does categorical perception in the left hemisphere depend on language. J. Exp. Psychol. Gen. 141, 439–443. Hugdahl, K., 2000. Lateralization of cognitive processes in the brain. Acta Psychol. 105, 211–235. Ivry, R.B., Robertson, L.C., 1998. The Two Sides of Perception. MIT Press, Cambridge, MA. Jager, G., Postma, A., 2003. On the hemispheric specialization for categorical and coordinate spatial relations: a review of the current evidence. Neuropsychologia 41, 504–515. Kemmerer, D., Tranel, D., 2000. A double dissociation between linguistic and perceptual representations of spatial relationships. Cogn. Neuropsychol. 17, 393–414. Kitterle, F., Christman, S., Hellige, J., 1990. Hemispheric differences are found in the identification, but not detection, of low vs. high spatial frequencies. Percept. Psychophys. 48, 297–306. Klatzky, R.L., 1998. Allocentric and egocentric spatial representations: definition, distinctions, and interconnections. Lect. Notes Comput. Sci. 1404, 1–17. Koenig, O., Reiss, L.P., Kosslyn, S.M., 1990. The development of spatial relation representations: evidence from studies of cerebral lateralization. J. Exp. Child Psychol. 50, 119–130. Kosslyn, S.M., 1987. Seeing and imagining in the cerebral hemispheres: a computational analysis. Psychol. Rev. 94, 148–175. Kosslyn, S.M., Koenig, O., Barrett, A., Backer Cave, C., Tang, J., Gabrieli, J.D.E., 1989. Evidence for two types of spatial representations: hemispheric specialization for categorical and coordinate relations. J. Exp. Psychol. Hum. 15, 723– 735. Kosslyn, S.M., Chabris, C.F., Marsolek, C.J., Koenig, O., 1992. Categorical versus coordinate spatial relations: computational analyses and computer simulations. J. Exp. Psychol. Hum. 181, 562–577. Kosslyn, S.M., Thompson, W.L., Gitelman, D.R., Alpert, N.M., 1998. Neural systems that encode categorical versus coordinate spatial relations: PET investigations. Psychobiology 26, 333–347. Kosslyn, S.M., 2006. You can play 20 questions with nature and win: categorical versus coordinate spatial relations as a case study. Neuropsychologia 44, 1519–1523. Kosslyn, S.M., Jacobs, R.A., 1994. Encoding shape and spatial relations: a simple mechanism for coordinating complementary representations. In: Honavar, V., Uhr, L.M. (Eds.), Artificial Intelligence and Neural Networks: Steps Toward Principled Integration. Academic Press, Boston. Laeng, B., 1994. Lateralization of categorical and coordinate spatial functions: a study of unilateral stroke patients. J. Cogn. Neurosci. 6, 189–203. Laeng, B., 2014. Representation of spatial relations. In: Ochsner, K., Kosslyn, S.M. (Eds.), The Oxford Handbook of Cognitive Neuroscience. Core Topics, vol. 1. Oxford University Press, Oxford, pp. 28–59. Laeng, B., Chabris, C., Kosslyn, S.M., 2003. Asymmetries in encoding spatial relations. In: Hugdahl, K., Davidson, R. (Eds.), The Asymmetrical Brain. MIT Press, Cambridge, MA. Laeng, B., Okubo, M., Saneyoshi, A., Michimata, C., 2011. Processing spatial relations with different apertures of attention. Cogn. Sci. 35 (2), 297–329. Laeng, B., Shah, J., Kosslyn, S.M., 1999. Identifying objects in conventional and contorted poses: contributions of hemisphere-specific mechanisms. Cognition 70, 53–85. Majid, A., Bowerman, M., Kita, S., Haun, D.B.M., Levinson, S.C., 2004. Can language restructure cognition? The case for space. Trends Cogn. Sci. 8 (3), 108–114. Marr, D., 1982. Vision: A Computational Investigation into the Human Representation and Processing of Visual Information. W.H. Freeman, San Francisco. Martin, M., 1979. Hemispheric specialization for local and global processing. Neuropsychologia 17, 33–40. Martin, R., Houssemand, C., Schiltz, C., Burnod, Y., Alexandre, F., 2008. Is there continuity between categorical and coordinate spatial relations coding: evidence from a grid/no-grid working memory paradigm. Neuropsychologia 46 (2), 576–594.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627
McNamara, T.P., 1986. Mental representations of spatial relations. Cognit. Psychol. 18, 87–121. Mecacci, L., 1993. On spatial frequencies and cerebral hemispheres: some remarks from the electrophysiological and neuropsychological points of view. Brain Cogn. 22, 199–212. Michimata, C., 1997. Hemispheric processing of categorical and coordinate spatial relations in vision and visual imagery. Brain Cogn. 33, 370–387. Michimata, C., Saneyoshi, A., Okubo, M., Laeng, B., 2011. Effects of the global and local attention on the processing of categorical and coordinate spatial relations. Brain Cogn. 77 (2), 292–297. Navon, D., 1977. Forest before trees: the precedence of global features in visual perception. Cognit. Psychol. 9, 353–383. Newell, A., 1973. You can’t play 20 questions with nature and win. In: Chase, W.G. (Ed.), Visual Information Processing. Academic Press, New York. Niebauer, C., Christman, S., 1999. Visual field differences in spatial frequency discrimination. Brain Cogn. 41, 381–389. Okubo, M., Michimata, C., 2002. Hemispheric processing of categorical and coordinate spatial relations in the absence of low spatial frequencies. J. Cogn. Neurosci. 14, 291–297. Okubo, M., Michimata, C., 2004. The role of high spatial frequencies in hemispheric processing of categorical and coordinate spatial relations. J. Cogn. Neurosci. 16, 1576–1582. Okubo, M., Laeng, B., Saneyoshi, A., Michimata, C., 2010. Exogenous attention differentially modulates the processing of categorical and coordinate spatial relations. Acta Psychol. 135 (1), 1–11. Paillard, J., 1991. Motor and representational framing of space. In: Paillard, J. (Ed.), Brain and Space. Oxford, UK, Oxford University Press, pp. 163–182. Palermo, L., Bureca, I., Matano, A., Guariglia, C., 2008. Hemispheric contribution to categorical and coordinate representational processes: a study on braindamaged patients. Neuropsychologia 46 (11), 2802–2807. Proverbio, A.M., Zani, A., Avella, C., 1997. Hemispheric asymmetries for spatial frequency discrimination in a selective attention task. Brain Cogn. 34, 311–320. Robertson, L.C., Lamb, M.R., 1991. Neuropsychological contributions to theories of part/whole organisation. Cognit. Psychol. 23, 299–330. Ruggiero, G., Ruotolo, F., Iachini, T., 2012. Egocentric/allocentric and coordinate/categorical haptic encoding in blind people. Cogn. Process. 13, S313–S317. Ruotolo, F., Postma, A., Iachini, T., van der Ham, I.J.M., 2011a. Frames of reference and categorical and coordinate spatial relations: a hierarchical organization. Exp. Brain Res. 214, 587–595. Ruotolo, F., van der Ham, I.J., Iachini, M., Postma, T.A., 2011b. The relationship between allocentric and egocentric frames of reference and categorical and coordinate spatial information processing. Q. J. Exp. Psychol. 64, 1138– 1156. Saneyoshi, A., Kaminaga, T., Michimata, C., 2006. Hemispheric processing of categorical/metric properties in object recognition. Neuroreport 17, 517–521.
7
Saneyoshi, A., Michimata, C., 2009. Lateralized effects of categorical and coordinate spatial processing of component parts on the recognition of 3-D non-namable objects. Brain Cogn. 71, 181–186. Schatz, J., Craft, S., Koby, M., DeBaun, M.R., 2004. Asymmetries in visual–spatial processing following childhood stroke. Neuropsychology 18, 340–352. Sereno, M.I., Huang, R.-S., 2006. A human parietal face area contains head-centered visual and tactile maps. Nat. Neurosci. 9, 1337–1343. Sergent, J., 1983. Role of input in visual hemispheric asymmetries. Psychol. Bull. 93, 481–512. Suegami, T., Laeng, B., 2013. A left cerebral hemisphere’s superiority in processing spatial-categorical information in a non-verbal semantic format. Brain Cogn. 81, 294–302. Trojano, L., Grossi, D., Linden, D.E.J., Formisano, E., Goebel, R., Cirillo, S., Elefante, R., Di Salle, F., 2002. Coordinate and categorical judgements in spatial imagery. An fMRI study. Neuropsychologia 40, 1666–1674. Trojano, L., Conson, M., Maffei, R., Grossi, D., 2006. Categorical and coordinate spatial processing in the imagery domain investigated by rTMS. Neuropsychologia 44, 1569–1574. van der Ham, I.J.M., Borst, G., 2011a. Individual differences in spatial relation processing: effects of strategy, ability, and gender. Brain Cogn. 76 (1), 184–190. van der Ham, I.J.M., Borst, G., 2011b. The nature of categorical and coordinate spatial processing: an interference study. J. Cognit. Psychol. 23 (8), 922–930. van der Ham, I.J.M., Dijkerman, H.C., van den Berg, E., 2012a. The effect of attentional scope on spatial relation processing: a case study. Neurocase 19 (5), 505–512. van der Ham, I.J.M., Duijndam, M.J.A., Raemaekers, M., van Wezel, R.J., Oleksiak, A., Postma, A.A., 2012b. Retinotopic mapping of categorical and coordinate spatial relation processing. PLoS ONE 7 (6), e38644. van der Ham, I.J.M., Oleksiak, A., van Wezel, R.J.A., van Zandvoort, M.J.E., Frijns, C.J.M., Kappelle, L.J., Postma, A., 2012c. The effect of stimulus features on working memory of categorical and coordinate spatial relations in patients with unilateral brain damage. Cortex 48, 737–745. van der Ham, I.J.M., Postma, A., 2010. Lateralization of spatial categories: a comparison of verbal and visuospatial categorical relations. Mem. Cognit. 38 (5), 582–590. van der Ham, I.J.M., van Strien, J.W., Oleksiak, A., van Wezel, R.J.A., Postma, A., 2010. Temporal characteristics of working memory for spatial relations: an ERP study. Int. J. Psychophysiol. 77, 83–94. van der Ham, I.J.M., van Zandvoort, M.J.E., Frijns, C.J.M., Kappelle, L.J., Postma, A., 2011. Hemispheric differences in spatial relation processing in a scene perception task: a neuropsychological study. Neuropsychologia 49, 999–1005. Van Kleeck, M., 1989. Hemispheric differences in global versus local processing of hierarchical visual stimuli by normal subjects: new data and a meta-analysis of previous studies. Neuropsychologia 27, 1165–1178. Wilkinson, D., Donnelly, N., 1999. The role of stimulus factors in making categorical and coordinate spatial judgments. Brain Cogn. 39, 171–185.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
ARTICLE IN PRESS
NBR 1952 1–7
Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Neuroscience and Biobehavioral Reviews journal homepage: www.elsevier.com/locate/neubiorev
Review
1
Lateralized perception: The role of attention in spatial relation processing
2
3
4 5 6 7
Q1
Ineke J.M. van der Ham a,∗ , Albert Postma a,b , Bruno Laeng c a b c
Department of Experimental Psychology, Utrecht University, The Netherlands Department of Neurology, Utrecht University Medical Center, The Netherlands Department of Psychology, University of Oslo, Norway
8
9 20
a r t i c l e
i n f o
a b s t r a c t
10 11 12 13 14
Article history: Received 16 December 2013 Received in revised form 8 May 2014 Accepted 13 May 2014
15
19
Keywords: Spatial relation processing Hemispheric lateralization Spatial attention
21
Contents
16 17 18
22 23 24 25 26 27 28 29 30
1. 2. 3. 4. 5. 6. 7.
Any spatial situation can be approached either categorically – the window is to my left – or coordinately – the glass is 20 cm away from the bottle. Since the first description of the distinction between categorical and coordinate spatial relation processing, it has often been shown that they are processed by at least partially different underlying mechanisms, mainly located in the left and right hemisphere, respectively. A number of recent studies have suggested that spatial attention plays a particularly important part in the perception of space: categorical processing benefits from a local focus of attention, and coordinate processing profits from a global focus of attention. This suggests that the lateralization pattern is modified by the concurrent size of the attentional focus, and is consequently more dynamic than previously thought. Therefore, a thorough revision of earlier theories on spatial relation processing is in order. In this review, we present a new model on lateralization of spatial relation processing that explicitly describes the role of spatial attention. © 2014 Published by Elsevier Ltd.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Empirical evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Critical evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Attentional scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bringing together dichotomies in low-level perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Generalizing the dichotomy to higher level perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
00 00 00 00 00 00 00 00 00
31 39 32
33 34 35 36 37 38
1. Introduction When we interact with our environment, its spatial features are crucial to us. There are different ways to deal with these spatial features. If someone asks us where we left the car keys, we can respond with something like: ‘I left them on the coffee table in the living room’. On the other hand, when we want to pick up those keys, it does not help much to know they are on the table; we need
∗ Corresponding author at: Heidelberglaan 1, Room H0.13, 3584 CS Utrecht, The Netherlands. Tel.: +31 030 2534023; fax: +31 030 2534511. E-mail address: [email protected] (I.J.M. van der Ham).
to know their exact location with respect to our hand in order to plan and execute our movement. These two examples illustrate the two main types of spatial relation processing. The first example shows that we can process spatial relations between objects in an abstract, propositional way, which is termed categorical. We can use this type of spatial relation to give directions or memorize where we left things. In contrast, in the second example we use the precise, metric properties of spatial relations, or coordinate spatial relations, when we navigate or grasp objects. The characteristics of these spatial relations are in part dictated by the type of ‘coordinate system’ that is used to represent spatial characteristics. Such coordinate systems can be either egocentric (observer based) or allocentric (environment based) (see
http://dx.doi.org/10.1016/j.neubiorev.2014.05.006 0149-7634/© 2014 Published by Elsevier Ltd.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
40 41 42 43 44 45 46 47 48 49 50 51 52
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
2
e.g. Klatzky, 1998; Majid et al., 2004). In an egocentric coordinate system, these spatial relations by definition concern the relation between an external object and the observer, such as ‘the car is to 55 my left’ or ‘I am closer to the library than you are’. Neurophysiolog56 ical studies on primates indicate that visual space is constructed 57 many times over, according to a variety of frames of reference, 58 each attached to different parts of the body and possibly subserv59 60Q2 ing different functions (Graziano and Gross, 1995). Neuroimaging in humans confirms that one of the topographic maps of the pari61 etal lobes is based on a head-centered coordinate frame (Sereno 62 and Huang, 2006). A plurality of sensorimotor action spaces can be 63 related to specific effectors that have the ability to move indepen64 dently from the rest of the body (e.g. hand, head, and eye). In such 65 motor-oriented frames of reference, spatial relationships between 66 two locations can be coded in terms of the movements required to 67 act or move from one to the other (Paillard, 1991). 68 In an allocentric coordinate system the relation concerns two 69 objects or parts of objects that are defined according to a refer70 ence frame whose fulcrum is located outside of the observer‘s body. 71 Examples of spatial relations from an allocentric perspective are: 72 ‘the church is to the North of the square’ and ‘you are 200 meters 73 West from the library’. Crucially, both categorical and coordinate 74 spatial relations can be applied within these different frame of ref75 erence and coordinate systems (see also Ruotolo et al., 2011a,b; 76 Ruggiero et al., 2012). In this review we focus on the direct com77 parison of categorical and coordinate spatial relations, regardless 78 of the coordinate system used. All that is discussed here can equally 79 apply to spatial relation processing in an egocentric as well as an 80 allocentric or object-based perspective. 81 The distinction between categorical and coordinate spatial rela82 tions is not only reflected by clear differences in functionality, 83 but also refers to different underlying processing mechanisms. 84 Although described earlier by McNamara (1986), Kosslyn (1987) 85 provides the first elaborate theoretical framework to understand 86 the differences between categorical and coordinate processing. He 87 introduced the idea that the two types of spatial relations are pro88 cessed by two different subsystems in the human brain, which have 89 lateralized due to evolutionary processes. Categorical spatial rela90 tion processing is thought to be lateralized to the left, whereas 91 coordinate spatial relation processing is argued to be right later92 alized. 93 Over the years, many experimental studies have been performed 94 concerning this distinction, which have enabled a fine-tuning of the 95 original theory (see for review, e.g. Jager and Postma, 2003; Laeng 96 et al., 2003; Laeng, 2014). Recent empirical studies have brought 97 new evidence to the table and point toward a substantial role of 98 spatial attention in explaining differences between categorical and 99 coordinate processing. This offers a novel theoretical perspective. 100 Therefore, an up-to-date review of current evidence and the impact 101 of these findings on the theoretical framework underlying cate102 gorical and coordinate spatial relation processing is called for. In 103 this review, we first provide an overview of the main experimental 104 work on spatial relation processing, particularly recent work. Then 105 we discuss what this tells us about the nature of spatial relation 106 processing, and focus on the role of spatial attention in particu107 lar. We present a new framework to fit the experimental findings 108 on how spatial attention affects spatial relation processing and its 109 lateralization. Lastly, we will explore how spatial relation theory 110 translates to human visual perception in general. 111 53 54
112
113 114 115
2. Empirical evidence Soon after Kosslyn’s first publication on the topic of spatial relation processing (Kosslyn, 1987) numerous behavioral studies attempted to empirically prove the distinction (e.g. Hellige and
Michimata, 1989; Kosslyn et al., 1989; Koenig et al., 1990; Bruyer et al., 1997; Wilkinson and Donnelly, 1999; Banich and Federmeier, 1999). These studies were mainly aimed at verifying the double dissociation of type of spatial relation and hemisphere. Typically, very simple stimuli were used in these studies. The dot-bar stimuli have been used in most of these early studies. These stimuli consist of a horizontal bar with a dot presented either above or below the bar, at varying distances, as illustrated in Fig. 1. The categorical instruction in this case, was to indicate whether the dot is ‘above’ or ‘below’ the bar, regardless of distance. The coordinate instruction in turn, was to focus on the distance and indicate whether the dot is ‘within’ or ‘not within’ a predetermined distance, regardless of side. An important feature of this design was that the stimulus layout was identical and only the instruction varied between the two versions of the task. In these behavioral studies, a visual half field design was used to assess lateralization patterns by means of response times and accuracy for stimuli presented to the left or right visual field. This approach was soon followed by studies in which lateralization was determined with more direct and precise measures of brain activity: positron emission tomography (PET), functional magnetic resonance, imaging (fMRI and MEG), electroencephalography (EEG), and transcranial magnetic stimulation (TMS). In the vast majority of these studies, the lateralization pattern was further substantiated (e.g. Baciu et al., 1999; Kosslyn et al., 1998; Trojano et al., 2002, 2006; van der Ham et al., 2009; Franciotti et al., Q3 2013). Several studies on patients with unilateral brain damage also supported these findings (e.g. Laeng, 1994; Palermo et al., 2008; van der Ham et al., 2011, 2012a,b,c). Overall, the parietal cortex Q4 seemed of particular importance; left parietal cortex was mainly linked to categorical processing, whereas the right parietal cortex was involved mostly with coordinate processing. Yet, the original theory as proposed by Kosslyn needed some attenuation. Most of the foregoing findings indicate that the lateralization pattern is not mutually exclusive: both hemispheres are involved in processing each type of spatial relations, but they show a clear bias toward one of the two types. Aside from the many techniques that have been used to study this dichotomy, various task designs have been applied as well, facilitating different research questions. The first number of studies was designed to study spatial relation processing in perception: a response was required to a briefly presented, simple stimulus. Results from these studies illustrate how we process the spatial features of visual stimuli. Later experiments also focused on spatial relation processing in working memory task designs, in which two sequentially presented stimuli were compared (e.g. Laeng, 1994; van der Ham et al., 2007), and in a mental imagery paradigm (Michimata, 1997; Palermo et al., 2008) where clock faces had to be imagined. These studies have shown that the typical lateralization pattern reflecting separate processing mechanisms can also be generalized to visual working memory and mental imagery. Although the double dissociation of relation type and hemisphere is found in the convincing majority of these studies, the effects have shown to be sensitive to the methodology used. For instance, Baciu et al. (1999) showed that over time, participants started to categorize the coordinate task, as reflected by a shift from a right hemisphere to a left hemisphere dominance during the experiment. This has been explained with a critical and disadvantageous feature of the dot-bar stimulus layout. The dot could only appear at a fixed number of positions and the instruction entailed a categorization of distances into ‘within’ a particular distance and ‘not within’ a particular distance. Therefore, participants could well start to realize this limitation and categorize the distances instead of encoding them as exact distances. The cross-dot working memory paradigm (see Fig. 2) in which two distances were compared later solved this problem as participants continued to encode the
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
3
Fig. 1. The often used dot-bar stimulus, (A) with all possible dot positions, (B) dots above the bar belonged to one spatial category, the dots below the bar to the other spatial category, and dots within a particular distance of the bar belonged to one type of coordinate relation, dots not within this distance belonged to the other type of coordinate relation. Note the during the task, only one dot position would be visible.
182 183 184 185 186 187 188 189 190 191 192
exact distance, not a distance category in this type of task. Also the higher amount of possible dot positions avoided categorization. Another study (van der Ham et al., 2007) showed that categorical information decays slower than coordinate information. In a working memory task design, the interval between two stimuli therefore determines whether a double dissociation is found or not, typically it is found with an interval of 500–2000 ms. In short, a large number of studies making use of various techniques and task designs demonstrate that there is a relative advantage of the left hemisphere in processing categorical information, and of the right hemisphere in processing coordinate
information. Yet, specific task demands can affect this lateralization pattern. 3. Critical evaluation A main point of criticism that can be made is whether it is really meaningful to study this distinction. What could we learn from such a distinction? Newell (1973) criticizes the tendency of psychologists to test binary distinctions, as supposedly these are typically not present in nature. Kosslyn (2006) argues against this statement, as for spatial relation processing such a clear distinction
Fig. 2. Version of the cross dot stimulus, used in working memory designs, (A) with all possible dot positions, (B) with a given dot position for the first stimulus (S1), the dots shown as S2 that would match categorically are shown on the left, and the dots that would match coordinately are shown to the right. Note that during the task, only one dot position would be visible for both S1 and S2.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
193 194
195
196 197 198 199 200 201
G Model NBR 1952 1–7 4
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
is present. In fact, the general idea behind lateralization of spatial relation processing should be viewed in a more general way. The brain is thought to ‘divide-and-conquer’ when it comes to processing information. If incoming information needs to be processed, it would be inefficient if both hemispheres would do this in the same way and then needed to communicate about this. Therefore, from an evolutionary point of view, lateralization, or the division of labor between the two hemispheres, will lead to more efficient and faster processing (see e.g. Hugdahl, 2000; Cook, 1986). Although categorical and coordinate spatial relations are generally thought to be based on partially separate processing mechanisms, sufficient evidence exists for it to be considered a clear binary, and lateralized distinction. Furthermore, we would like to argue that studying how we process spatial relations is meaningful, as it can be generalized to many areas of perception, as discussed later. Despite the large amount of supporting evidence for the dichotomy, alternative explanations have been proposed as well. One prevalent alternative is that of space versus language. The traditional line of reasoning is that the right hemisphere typically processes spatial information, whereas verbal information processing is located in the left hemisphere. Therefore, it has been argued that in particular the left hemisphere advantage of categorical processing is due to the linguistic properties of this type of processing. In the dot bar task for instance, the use of propositions ‘above’ and ‘below’ suffices to solve the task. Kemmerer and Tranel (2000) argued there might be a triad instead of a dichotomy, with categorical relations split up into strictly spatial categories and verbal categories. They proposed that only the processing of verbal categories shows a left hemisphere bias. Yet, a number of studies argue against the fact that left lateralization is based primarily on the verbal content of processing. For categorical perception in general, it has been shown that preverbal infants show mainly left hemisphere activity when they categorize spatial orientations (Franklin et al., 2010). Also, no difference in lateralization pattern is found when people use labeled or unlabeled categories, indicating that language is not the main factor in the direction of lateralization (Holmes and Wolff, 2012). Within spatial relation processing in particular, the verbal nature of categorical stimuli could only explain for a minor part of the left lateralization (van der Ham and Postma, 2010). This is further substantiated by Suegami and Laeng (2013), who found that although categorical relations can be coded semantically as well as visuospatially, a left lateralization is found for both. As language processing is convincingly based in the left hemisphere, it is likely that it has some effect on the extent of lateralization when participants use verbal strategies to solve categorical tasks. Yet, given the current evidence it seems highly unlikely that language by itself is the determining factor in the direction of lateralization. A second alternative explanation that has come up is that of difficulty. Typically, processing categorical relations is easier to do than processing coordinate relations. Sergent (1991a,b) was the first to therefore suggest that instead of a dichotomy of categorical and coordinate, it might be a continuum between easy and difficult. The more difficult a task would get, the more the right hemisphere would get involved. Although Sergent’s point of view was experimentally disproven by Kosslyn et al. (1992) and Slotnick et al. (2001), this point has also been used to explain experimental results by Van der Lubbe et al. (2006) and Martin et al. (2008) in later studies. Yet, a more recent fMRI study van der Ham et al. (2012a,b,c) again demonstrated that task difficulty itself cannot explain for the lateralization patterns found (see also Franciotti et al., 2013). It cannot be denied that there are adherent difficulty differences between categorical and coordinate processing, but when reviewing the evidence, difficulty does not seem to determine the lateralization pattern.
4. Attentional scope Given that categorical and coordinate spatial relation processing are markedly distinct at a neurocognitive level and cannot be merely explained by characteristics like involvement of language or task difficulty, the question remains what exactly is the nature of spatial relation processing? A number of recent studies suggest a novel, appealing answer to this question, by looking into spatial attention and more specifically the size of attentional scope. Spatial attention is not an entirely new element in the ideas on spatial relation processing. Actually, in his first report Kosslyn (1987) already commented on the issue. He described the purpose of spatial relation interpretation as looking up the location of the relevant part and directing attention to this location, leading to the output of instructing the attention-shift subsystem to shift attention accordingly. He proposed that for both categorical and coordinate interpretation this process is the same, except for the nature of the representation used to do this, which is either categorical or coordinate. Several behavioral studies indicate that the difference in representation – categorical or coordinate – is closely related to the size of the attended location. Categorical processing is found to benefit most from a local, small focus, whereas coordinate processing seems more efficient with a global, large focus of attention. In several experiments this same pattern was found when manipulating the area participants attended to (Okubo et al., 2010; Michimata et al., 2011; Borst and Kosslyn, 2010; Laeng et al., 2011; Franconeri et al., 2012; Franciotti et al., 2013). Moreover, the same attentional pattern was also found in the absence of experimental manipulations. In a case study, patient NC showed great difficulty in processing coordinate information in particular, which coincided with a clear local bias, as opposed to the global bias generally found in human visual perception (van der Ham et al., 2012a,b,c). In line with this, it has also been shown that children with right hemisphere damage show a combination of impairment in global focus and coordinate judgments (Schatz et al., 2004). Furthermore, with retinotopic mapping of fMRI data, it has been shown that participants directed their attention more locally when they were asked to give categorical responses, as compared to coordinate responses (van der Ham et al., 2012a,b,c). The interpretation of this dissociation between categorical and coordinate processing in terms of size of attentional focus revives an older idea about spatial relation processing: the link to receptive field sizes. In a computational model (Kosslyn et al., 1992) the flow of information in processing categorical and coordinate information was simulated. This simulation showed that the categorical decisions benefited from the use of small receptive fields, whereas the opposite was found for coordinate information. This difference makes sense when considering the functionality of the two types of relations. The smaller, non-overlapping fields facilitate splitting up space into distinct areas, or categories, and large, overlapping fields facilitate ‘coarse coding’ (Hinton, 1981), which helps to derive precise metric information. This is substantiated by the studies showing that attentional focus is different between categorical and coordinate processing. An important addition provided by the spatial attention studies is that they show that it concerns a dynamic process: the size of the attentional focus seems to be attenuated according to the task at hand, rather than to be fixed (Fig. 3). Based on the recent empirical evidence concerning the role of attentional focus, we present a new model to explain for the lateralization effects found for spatial relation processing. This model highlights that both task instruction and stimulus size have a pivotal role in predicting the lateralization effect. Task instruction reflects the original theory: a categorical task instruction will lead to a left hemisphere advantage, and a coordinate instruction to a right hemisphere advantage. Stimulus size has a major role in this
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
268
269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
Fig. 3. The model of information processing resulting in a spatial relation decision. Both task instruction and stimulus size contribute to the lateralization pattern.
359
model. It refers to the size of the crucial elements of the visual stimulus and directly determines the attentional focus that is required for a specific spatial relations decision. When stimulus size is relatively small, the attentional focus will also be relatively small and lead to a left hemisphere advantage in processing. Conversely, a relatively large stimulus size will create a large area of attentional focus and consequently result in a right hemisphere advantage. This new model highlights that lateralization in visual perception is not solely determined by task instruction. Attentional focus has been demonstrated to yield a considerable impact as well. The addition of stimulus size not only explains for the recent findings concerning small and large attentional scope and categorical and coordinate spatial processing, respectively, it could also account for variations in extent of lateralization found in different studies over the years. Precise stimulus features like size have been largely ignored over the years. The main criterion has been that the stimulus layout would be identical for categorical and coordinate decisions, in the sense that only the instruction would determine whether it was a categorical or coordinate task. Yet, the elements of those stimuli that are crucial for making a correct decision differ nonetheless. For instance, with the cross dot stimuli in Fig. 2, the categorical decision only requires to check one fourth of the total stimulus area (the second dot is either in that area or not), whereas for the coordinate decision, the entire stimulus area should be attended (the second dot can appear in any direction relative to the center). Therefore, it is important to realize there are two forces at work here.
360
5. Bringing together dichotomies in low-level perception
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358
5
spatial frequency hypothesis; the left hemisphere is more proficient in processing high frequency and the right hemisphere in processing low frequency information (Sergent, 1983). This has been supported by a number of studies (e.g. Mecacci, 1993; Proverbio et al., 1997). Surprisingly little research has been done in which these different dichotomies have been directly integrated. Frequency manipulations have been applied in categorical-coordinate tasks to investigate a possible interaction. Okubo and Michimata illustrated that the coordinate right hemisphere advantage disappeared when low spatial frequencies were eliminated from the stimuli (Okubo and Michimata, 2002, 2004). The categorical left hemisphere advantage disappeared when high spatial frequencies were filtered out in the stimuli. Yet, such a relation between spatial frequency and spatial relation processing is contradicted by Niebauer and Christman (1999). The ‘double filtering by frequency model’ created by Ivry and Q5 Robertson (1998) provides a theoretical framework for this particular topic. In their model they provide a unitary explanation of lateralization effects in visual perception. They base their claim on the premise that some of our perceptual systems make use of a Fourier analysis to describe incoming information, and that the left and right hemisphere have a bias toward processing high and low spatial frequencies, respectively. Their model is therefore based on the frequency characteristics of incoming stimuli. It proposes three separate and sequential stages of stimulus processing: sensory representation, selective filtering of task-relevant information, and asymmetric filtering by cerebral hemisphere of selected information. In the first filtering stage an attentional mechanism serves as a filtering system with regard to the frequency spectrum; the region selected in the spectrum can be as narrow or broad as required by the task performed. The actual lateralization does not occur until the second filtering stage, during which higher order analyses are taking place, for which each hemisphere has its preference. The processing in the left and right hemisphere is considered high-pass and low-pass filtering operations, respectively. As Ivry and Robertson (1998) state: “The central hypothesis of the DFF theory is that the output of those filtering operations underlies many of the laterality effects reported in visual and auditory perception”. These laterality effects are very likely to include local-global and categorical-coordinate processing. The authors indicate the model is a post hoc interpretation of experimental studies and new empirical input is necessary to examine its validity. The recent findings concerning attentional focus, described above appear to fulfill this need. As the local and global focus of spatial attention is directly related to categorical and coordinate spatial relation processing, respectively, these findings underline this model. Importantly, ‘task instruction’ as defined in the model does not seem to be restricted to spatial relation task instruction, but can easily be replaced by other types of spatial instructions, like local versus global.
6. Generalizing the dichotomy to higher level perception 361 362 363 364 365 366 367 368 369 370 371 372 373 374
By discussing size of attentional focus, the local versus global dichotomy naturally comes to mind. This is another classical dichotomy proposed to describe human perception (e.g. Navon, 1977) that has been shown to be lateralized (e.g. Martin, 1979). Although it is an inherent continuous property, it is typically regarded as a dichotomy of ‘relatively small’ versus ‘relatively large’. A meta-analysis confirmed the general finding that there is left hemispheric specialization for local information and the right hemisphere shows an advantage in processing global information (Van Kleeck, 1989). Furthermore, the notion of receptive field sizes is directly related to the processing of high and low spatial frequencies, also a dichotomy with a distinct lateralization pattern. The commonly found direction of lateralization in spatial frequency processing tasks has been framed in the
With a new model reflecting the main findings concerning spatial relation processing, it is imperative to attempt to explore generalization of this theory. As mentioned earlier, the vast majority of fundamental studies providing information about how we process categorical and coordinate information use very simple stimulus layouts, consisting of lines, crosses, and dots. How do these findings relate to how we process visual information in real life? Several researchers have asked this question and looked into spatial relation processing within objects (e.g. Laeng et al., 1999; Saneyoshi et al., 2006; Saneyoshi and Michimata, 2009), faces (Cooper and Wojan, 2000), natural scenes (van der Ham et al., 2012a,b,c), and even in the dynamic process of navigation (Baumann et al., 2012). These studies have largely underlined the evidence from the
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423
424
425 426 427 428 429 430 431 432 433 434 435 436 437
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
6
studies with simple stimuli: a clear dissociation between the two types in terms of underlying processes. So, the distinction appears to hold regardless of visual complexity, meaning, and context. Yet, 440 Q6 in a study on spatial relations in objects Amorapanth et al. (2010) 441 apply some nuance to this: only part of the brain areas involved 442 in spatial relation processing were separate, other areas appeared 443 to be shared for both types. Taken together, it seems that the way 444 in which we perceive the spatial relation information in the world 445 finds its basis in the theories discussed here. 446 438 439
447
7. Conclusion
476
Over 25 years of research supports the view that categorical and coordinate spatial relations are processed by at least partially different underlying mechanisms, mainly located in the left and right hemisphere, respectively. Moreover, this distinction is found within perception (of either space or objects), working memory, and mental imagery. Several criticisms have been uttered in response to these findings. Yet, the alternative explanations for the findings recapitulated here do not match up to the existing theory: language only seems to play a minor part and a difficulty continuum seems highly unlikely. Spatial attention plays a particularly important part in recent empirical studies on the perception of space. Categorical processing benefits from a local focus of attention, whereas coordinate processing is more efficient when a global focus of attention is applied. Therefore, the original theory on the role of receptive field sizes has been revived, and seems to be more flexible than previously thought. Furthermore, it also highlights the interplay between various dichotomies widely described within perception and in part supports Ivry and Robertson’s Double Filtering by Frequency Model. We propose that the clear distinction between categorical and coordinate is closely linked to the continuous measure of attentional scope. The rigid distinction between the two can therefore be attenuated by this factor, as a result of experimental characteristics and innate preferences. Finally, we see that the findings discussed here do not only reflect how we process simple laboratory stimuli, but also how we see the world. The dichotomy has shown to be a major factor as well in how we process objects, faces, scenes, and interact with our environment.
477 Q7
Uncited references
448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475
478 479 480
481
482 Q8 483
484
485 486 487 488 489 490 491 492 493 494 495 496 497
Dent (2009), Kitterle et al. (1990), Kosslyn and Jacobs (1994), Marr (1982), Robertson and Lamb (1991), van der Ham and Borst (2011a,b) and van der Ham et al. (2010). Acknowledgment IH was supported by a Veni grant (451-12-004) by NWO (the Netherlands Organisation of Scientific Research). References Amorapanth, P., Widick, P., Chatterjee, A., 2010. The neural basis for spatial relations. J. Cogn. Neurosci. 22 (8), 1739–1753. Baciu, M., Koenig, O., Vernier, M.-P., Bedoin, N., Rubin, C., Segebarth, C., 1999. Categorical and coordinate spatial relations: fMRI evidence for hemispheric specialization. Neuroreport 10, 1373–1378. Banich, M.T., Federmeier, K.D., 1999. Categorical and metric spatial processes distinguished by task demands and practice. J. Cogn. Neurosci. 11, 153–166. Baumann, O., Chan, E., Mattingley, J.B., 2012. Distinct neural networks underlie encoding of categorical and coordinate spatial relations during active navigation. Neuroimage 60 (3), 1630–1637. Borst, G., Kosslyn, S.M., 2010. Varying the scope of attention alters the encoding of categorical and coordinate spatial relations. Neuropsychologia 48 (9), 2769–2772.
Bruyer, R., Scailquin, J.-C., Coibion, P., 1997. Dissociation between categorical and coordinate spatial computations: modulation by the cerebral hemispheres, task properties, mode of response, and age. Brain Cogn. 33, 245–277. Cook, N.D., 1986. The Brain Code: Mechanisms of Information Transfer and the Role of the Corpus Callosum. Methuen, London. Cooper, E.E., Wojan, T.J., 2000. Differences in the coding of spatial relations in face identification and basic-level object recognition. J. Exp. Psychol. Learn. 26, 470–488. Dent, K., 2009. Coding categorical and coordinate spatial relations in visual–spatial short-term memory. Q. J. Exp. Psychol. 62 (12), 2372–2387. Franciotti, R., D’Ascenzo, S., Di Domenico, A., Tommasi, L., Laeng, B., 2013. Focusing narrowly or broadly attention when judging categorical and coordinate spatial relations: a MEG study. PLoS ONE 8 (12), e83434. Franconeri, S.L., Scimeca, J.M., Roth, J.C., Helseth, S.A., Kahn, L.E., 2012. Flexible visual processing of spatial relationships. Cognition 122, 210–227. Franklin, A., Catherwood, D., Alvarez, J., Axelsson, E., 2010. Hemispheric asymmetries in categorical perception of orientation in infants and adults. Neuropsychologia 48, 2648–2657. Graziano, M.S.A., Gross, C.G., 1995. Multiple representations of space in the brain. Neuroscientist 1, 43–50. Hellige, J.B., Michimata, C., 1989. Categorization versus distance: hemispheric differences for processing spatial information. Mem. Cognit. 17, 770–776. Hinton, G.E., 1981. Shape representation in parallel systems. In: Drina, A. (Ed.), Proceedings of the Seventh International Joint Conference on Artificial Intelligence. MIT Press, Cambridge, MA. Holmes, K.J., Wolff, P., 2012. Does categorical perception in the left hemisphere depend on language. J. Exp. Psychol. Gen. 141, 439–443. Hugdahl, K., 2000. Lateralization of cognitive processes in the brain. Acta Psychol. 105, 211–235. Ivry, R.B., Robertson, L.C., 1998. The Two Sides of Perception. MIT Press, Cambridge, MA. Jager, G., Postma, A., 2003. On the hemispheric specialization for categorical and coordinate spatial relations: a review of the current evidence. Neuropsychologia 41, 504–515. Kemmerer, D., Tranel, D., 2000. A double dissociation between linguistic and perceptual representations of spatial relationships. Cogn. Neuropsychol. 17, 393–414. Kitterle, F., Christman, S., Hellige, J., 1990. Hemispheric differences are found in the identification, but not detection, of low vs. high spatial frequencies. Percept. Psychophys. 48, 297–306. Klatzky, R.L., 1998. Allocentric and egocentric spatial representations: definition, distinctions, and interconnections. Lect. Notes Comput. Sci. 1404, 1–17. Koenig, O., Reiss, L.P., Kosslyn, S.M., 1990. The development of spatial relation representations: evidence from studies of cerebral lateralization. J. Exp. Child Psychol. 50, 119–130. Kosslyn, S.M., 1987. Seeing and imagining in the cerebral hemispheres: a computational analysis. Psychol. Rev. 94, 148–175. Kosslyn, S.M., Koenig, O., Barrett, A., Backer Cave, C., Tang, J., Gabrieli, J.D.E., 1989. Evidence for two types of spatial representations: hemispheric specialization for categorical and coordinate relations. J. Exp. Psychol. Hum. 15, 723– 735. Kosslyn, S.M., Chabris, C.F., Marsolek, C.J., Koenig, O., 1992. Categorical versus coordinate spatial relations: computational analyses and computer simulations. J. Exp. Psychol. Hum. 181, 562–577. Kosslyn, S.M., Thompson, W.L., Gitelman, D.R., Alpert, N.M., 1998. Neural systems that encode categorical versus coordinate spatial relations: PET investigations. Psychobiology 26, 333–347. Kosslyn, S.M., 2006. You can play 20 questions with nature and win: categorical versus coordinate spatial relations as a case study. Neuropsychologia 44, 1519–1523. Kosslyn, S.M., Jacobs, R.A., 1994. Encoding shape and spatial relations: a simple mechanism for coordinating complementary representations. In: Honavar, V., Uhr, L.M. (Eds.), Artificial Intelligence and Neural Networks: Steps Toward Principled Integration. Academic Press, Boston. Laeng, B., 1994. Lateralization of categorical and coordinate spatial functions: a study of unilateral stroke patients. J. Cogn. Neurosci. 6, 189–203. Laeng, B., 2014. Representation of spatial relations. In: Ochsner, K., Kosslyn, S.M. (Eds.), The Oxford Handbook of Cognitive Neuroscience. Core Topics, vol. 1. Oxford University Press, Oxford, pp. 28–59. Laeng, B., Chabris, C., Kosslyn, S.M., 2003. Asymmetries in encoding spatial relations. In: Hugdahl, K., Davidson, R. (Eds.), The Asymmetrical Brain. MIT Press, Cambridge, MA. Laeng, B., Okubo, M., Saneyoshi, A., Michimata, C., 2011. Processing spatial relations with different apertures of attention. Cogn. Sci. 35 (2), 297–329. Laeng, B., Shah, J., Kosslyn, S.M., 1999. Identifying objects in conventional and contorted poses: contributions of hemisphere-specific mechanisms. Cognition 70, 53–85. Majid, A., Bowerman, M., Kita, S., Haun, D.B.M., Levinson, S.C., 2004. Can language restructure cognition? The case for space. Trends Cogn. Sci. 8 (3), 108–114. Marr, D., 1982. Vision: A Computational Investigation into the Human Representation and Processing of Visual Information. W.H. Freeman, San Francisco. Martin, M., 1979. Hemispheric specialization for local and global processing. Neuropsychologia 17, 33–40. Martin, R., Houssemand, C., Schiltz, C., Burnod, Y., Alexandre, F., 2008. Is there continuity between categorical and coordinate spatial relations coding: evidence from a grid/no-grid working memory paradigm. Neuropsychologia 46 (2), 576–594.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583
G Model NBR 1952 1–7
ARTICLE IN PRESS I.J.M. van der Ham et al. / Neuroscience and Biobehavioral Reviews xxx (2014) xxx–xxx
584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627
McNamara, T.P., 1986. Mental representations of spatial relations. Cognit. Psychol. 18, 87–121. Mecacci, L., 1993. On spatial frequencies and cerebral hemispheres: some remarks from the electrophysiological and neuropsychological points of view. Brain Cogn. 22, 199–212. Michimata, C., 1997. Hemispheric processing of categorical and coordinate spatial relations in vision and visual imagery. Brain Cogn. 33, 370–387. Michimata, C., Saneyoshi, A., Okubo, M., Laeng, B., 2011. Effects of the global and local attention on the processing of categorical and coordinate spatial relations. Brain Cogn. 77 (2), 292–297. Navon, D., 1977. Forest before trees: the precedence of global features in visual perception. Cognit. Psychol. 9, 353–383. Newell, A., 1973. You can’t play 20 questions with nature and win. In: Chase, W.G. (Ed.), Visual Information Processing. Academic Press, New York. Niebauer, C., Christman, S., 1999. Visual field differences in spatial frequency discrimination. Brain Cogn. 41, 381–389. Okubo, M., Michimata, C., 2002. Hemispheric processing of categorical and coordinate spatial relations in the absence of low spatial frequencies. J. Cogn. Neurosci. 14, 291–297. Okubo, M., Michimata, C., 2004. The role of high spatial frequencies in hemispheric processing of categorical and coordinate spatial relations. J. Cogn. Neurosci. 16, 1576–1582. Okubo, M., Laeng, B., Saneyoshi, A., Michimata, C., 2010. Exogenous attention differentially modulates the processing of categorical and coordinate spatial relations. Acta Psychol. 135 (1), 1–11. Paillard, J., 1991. Motor and representational framing of space. In: Paillard, J. (Ed.), Brain and Space. Oxford, UK, Oxford University Press, pp. 163–182. Palermo, L., Bureca, I., Matano, A., Guariglia, C., 2008. Hemispheric contribution to categorical and coordinate representational processes: a study on braindamaged patients. Neuropsychologia 46 (11), 2802–2807. Proverbio, A.M., Zani, A., Avella, C., 1997. Hemispheric asymmetries for spatial frequency discrimination in a selective attention task. Brain Cogn. 34, 311–320. Robertson, L.C., Lamb, M.R., 1991. Neuropsychological contributions to theories of part/whole organisation. Cognit. Psychol. 23, 299–330. Ruggiero, G., Ruotolo, F., Iachini, T., 2012. Egocentric/allocentric and coordinate/categorical haptic encoding in blind people. Cogn. Process. 13, S313–S317. Ruotolo, F., Postma, A., Iachini, T., van der Ham, I.J.M., 2011a. Frames of reference and categorical and coordinate spatial relations: a hierarchical organization. Exp. Brain Res. 214, 587–595. Ruotolo, F., van der Ham, I.J., Iachini, M., Postma, T.A., 2011b. The relationship between allocentric and egocentric frames of reference and categorical and coordinate spatial information processing. Q. J. Exp. Psychol. 64, 1138– 1156. Saneyoshi, A., Kaminaga, T., Michimata, C., 2006. Hemispheric processing of categorical/metric properties in object recognition. Neuroreport 17, 517–521.
7
Saneyoshi, A., Michimata, C., 2009. Lateralized effects of categorical and coordinate spatial processing of component parts on the recognition of 3-D non-namable objects. Brain Cogn. 71, 181–186. Schatz, J., Craft, S., Koby, M., DeBaun, M.R., 2004. Asymmetries in visual–spatial processing following childhood stroke. Neuropsychology 18, 340–352. Sereno, M.I., Huang, R.-S., 2006. A human parietal face area contains head-centered visual and tactile maps. Nat. Neurosci. 9, 1337–1343. Sergent, J., 1983. Role of input in visual hemispheric asymmetries. Psychol. Bull. 93, 481–512. Suegami, T., Laeng, B., 2013. A left cerebral hemisphere’s superiority in processing spatial-categorical information in a non-verbal semantic format. Brain Cogn. 81, 294–302. Trojano, L., Grossi, D., Linden, D.E.J., Formisano, E., Goebel, R., Cirillo, S., Elefante, R., Di Salle, F., 2002. Coordinate and categorical judgements in spatial imagery. An fMRI study. Neuropsychologia 40, 1666–1674. Trojano, L., Conson, M., Maffei, R., Grossi, D., 2006. Categorical and coordinate spatial processing in the imagery domain investigated by rTMS. Neuropsychologia 44, 1569–1574. van der Ham, I.J.M., Borst, G., 2011a. Individual differences in spatial relation processing: effects of strategy, ability, and gender. Brain Cogn. 76 (1), 184–190. van der Ham, I.J.M., Borst, G., 2011b. The nature of categorical and coordinate spatial processing: an interference study. J. Cognit. Psychol. 23 (8), 922–930. van der Ham, I.J.M., Dijkerman, H.C., van den Berg, E., 2012a. The effect of attentional scope on spatial relation processing: a case study. Neurocase 19 (5), 505–512. van der Ham, I.J.M., Duijndam, M.J.A., Raemaekers, M., van Wezel, R.J., Oleksiak, A., Postma, A.A., 2012b. Retinotopic mapping of categorical and coordinate spatial relation processing. PLoS ONE 7 (6), e38644. van der Ham, I.J.M., Oleksiak, A., van Wezel, R.J.A., van Zandvoort, M.J.E., Frijns, C.J.M., Kappelle, L.J., Postma, A., 2012c. The effect of stimulus features on working memory of categorical and coordinate spatial relations in patients with unilateral brain damage. Cortex 48, 737–745. van der Ham, I.J.M., Postma, A., 2010. Lateralization of spatial categories: a comparison of verbal and visuospatial categorical relations. Mem. Cognit. 38 (5), 582–590. van der Ham, I.J.M., van Strien, J.W., Oleksiak, A., van Wezel, R.J.A., Postma, A., 2010. Temporal characteristics of working memory for spatial relations: an ERP study. Int. J. Psychophysiol. 77, 83–94. van der Ham, I.J.M., van Zandvoort, M.J.E., Frijns, C.J.M., Kappelle, L.J., Postma, A., 2011. Hemispheric differences in spatial relation processing in a scene perception task: a neuropsychological study. Neuropsychologia 49, 999–1005. Van Kleeck, M., 1989. Hemispheric differences in global versus local processing of hierarchical visual stimuli by normal subjects: new data and a meta-analysis of previous studies. Neuropsychologia 27, 1165–1178. Wilkinson, D., Donnelly, N., 1999. The role of stimulus factors in making categorical and coordinate spatial judgments. Brain Cogn. 39, 171–185.
Please cite this article in press as: van der Ham, I.J.M., et al., Lateralized perception: The role of attention in spatial relation processing. Neurosci. Biobehav. Rev. (2014), http://dx.doi.org/10.1016/j.neubiorev.2014.05.006
628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
